Image Processing Reference
In-Depth Information
(a)
(b)
(c)
FIGURE 6.3
(a) Blood vessel image, (b) intuitionistic fuzzy method by Couto and (c) intuitionistic fuzzy
method by Bustince.
Example 6.3
An example of a biomedical image is shown in Figure 6.3, which uses
intuitionistic fuzzy entropy methods. Blood vessel images are very
poorly illuminated, and thresholding becomes very difficult. Blood ves-
sel thresholding is very important in pathology when there is a need
to count the number of blood vessels in diagnosing diseases such as
prostate cancer. Figure 6.3b is the intuitionistic fuzzy entropy method
by Couto, and Figure 6.3c is the intuitionistic fuzzy entropy method by
Bustince.
6.4.2 Intuitionistic Fuzzy Divergence-Based Method
Vlachos and Sergiadis [24] suggested an intuitionistic fuzzy image thresh-
olding method that is similar to the idea as described by Chaira and Ray
[5] where the minimization of actual and ideal thresholded image leads to
maximum belongingness of foreground pixels to foreground regions and
background pixels to background regions.
The gamma membership function is calculated from the gamma distribu-
tion [9,10], which is given as
f
(
x
) = exp(−(
x
−
m
))
(6.11)
Let us consider an image
A
of size
M
×
M
with maximum grey level
L
and
g
ij
the (
i
,
j
)th pixel of the image with
i
,
j
= 1, 2, 3, …,
M
. Then, given a certain
threshold '
t
' that separates the objects and the background regions, the
Search WWH ::
Custom Search